What to Do When a Client Doesn't Realize the Impact of Your Work

New client, three weeks in, ads are crushing. But the client is skeptical.

If you've been reading my blog for a bit, you'll remember me mentioning this client two weeks ago. The one I closed after 7 months of “soft” follow-ups.

Well, so far so good in terms of delivering quality work.

I'm sitting in front of an ad account that's beating his previous baseline on every metric I can pull. Cost per lead is down, CPC is down, and conversion rate is up.

Yet, the client is skeptical.

Not "let's see how next week goes" skeptical. More like "I'm not sure this is working" skeptical. Reaching for reasons it might not be real.

If you've been freelancing long enough, you’ve come across a client or two like that.

One where you REALLY need the data to help you make a case for the impact of your work.

So that's what we're talking about today.

‍

The Client Isn't The Problem

Before I go any further, I want to say something I had to remind myself of last week.

I'm not mad at this client.

He's three weeks into a new retainer with a new freelancer running a strategy he's never tried, with his money on the line. Of course he's nervous. He should be. He's not a data nerd. He doesn't live in Meta Ads Manager the way I do. He's running a business.

His skepticism doesn't make sense looking at the numbers.

But it makes total sense looking at him.

And once I clocked that, the whole problem reshaped itself.

He wasn't telling me the work wasn't working. He was telling me he couldn't see that it was.

Different problem. Different solution.

‍

Why Measurable Work Saves You

Here's where this stops being a story and starts being a lesson.

The reason I could even have this conversation, the reason I'm not currently spiraling into a "did I just sign a client who's about to fire me" panic, is because I sold a service I could measure.

CTR, CPA, CPL, account spend, % new customers. The agreement we signed has KPIs baked in. We agreed on the baseline. We agreed on what "working" looks like.

So when the client got emotional, I had something to point at.

If I'd sold "brand strategy" or "marketing support" or any of the vague stuff that lives in the deck and nowhere else, this same conversation would have ended very differently. Probably with a Slack message I don't want to imagine.

Sell vague work, and you've got nothing to defend yourself with when the client's mood turns.

Sell measurable work, and the numbers do the arguing for you.

That's the whole game, especially in the first 30 days.

‍

What This Looks Like In Practice

Five things I do (or wish I'd done sooner) so this conversation goes how it just did, instead of how it could have.

1. Sell only services you can tie to clean KPIs

If you can't measure it, you can't defend it.

When you're scoping the engagement, pick KPIs the client agrees to before you start. Not aspirational ones. Real ones the platform actually reports. CTR, CPA, ROAS for ads. Open rate, click rate, revenue per send for email. Whatever the channel is, write the metrics into the SOW.

This sounds tedious. It's the most important 20 minutes of any new engagement.

2. Lock the baseline on day one

Before you push anything live, pull the last 60 to 90 days of their data.

Their old creative did this CTR. Their old funnel did this CPA. Their old whatever did this whatever.

Now you have a wall to beat. And more importantly, you have a number to point at when the client says "I'm not sure if this is better."

3. Build the proof loop into the deliverable

Every week, pull the comparison. Old creative vs new. Pre-engagement vs post. Show the deltas. Translate them into business language.

Don't wait for the client to ask. By the time they ask, they're already emotional. The proof loop has to land before the doubt does.

4. Translate the numbers into plain English

Your client doesn't live in the ad account. They don't know what a 1.8 ROAS means in context. They don't remember if 2.3% CTR is good or bad for their category.

Your job isn't to deliver the data. It's to translate it.

"Last month's creative was getting clicks at $4.20. Yours is getting them at $2.80. That's a 33% drop. Same budget, 50% more clicks."

That sentence does more work than a 12-tab spreadsheet.

5. When the doubt shows up, let the data argue

This is the one most freelancers screw up. You get pushback, your ego kicks in, and you write a long email defending yourself.

Don't.

Pull the comparison. Walk through it line by line. Let the numbers do the convincing. The data is neutral. You aren't.

The conversation I had with this client last week didn't end with him saying "you're right and I'm wrong." It ended with him saying "what should we test next?"

That's the only win that matters.

‍

The Old Freelancer vs The Modern Freelancer

The old freelancer sells "marketing services."

Vague scope. Vague deliverables. Vague KPIs (if there are any at all). And when the client gets cold feet at week 3, they're left arguing the relationship. "Trust me, this is going to work." "Give it more time." "I've done this before."

The modern freelancer sells measurable outcomes.

Real KPIs in the SOW. A baseline locked on day one. A weekly proof loop built into the work. So when the emotional moment shows up (and it always shows up) the numbers are already on the table doing the work.

‍

Your Action Step

Look at your current proposal template, or your last SOW.

Find the section where you describe what you're going to deliver.

Underneath it, add three lines:

Baseline metric we're starting from.

Target metric we're aiming for.

How often I'll report on both.

If your current scope can't answer those three questions, your scope is too vague. Tighten it before you send the next one.

That's it. Watch how different the first 30 days feels with the next new client you sign.

Sell something you can measure.

Then let the numbers fight for you.

‍

Want more freelance marketing tips that actually work?

Subscribe to my newsletter and get actionable freelance marketing strategies delivered to your inbox weekly.